Stereo images have to be calibrated before stereo vision can recover the 3D information of the imaged scene. Position constraints via point correspondences are traditionally used to solve the calibration problem. We d...
详细信息
Stereo images have to be calibrated before stereo vision can recover the 3D information of the imaged scene. Position constraints via point correspondences are traditionally used to solve the calibration problem. We describe a method that uses correspondences of projections of orthogonal trihedral vertices for calibration. the method has a closed-form solution, as opposed to many other calibration methods which are iterative. It also requires only two vertex correspondences at minimum to recover all the transformation parameters which are recoverable from a stereo image pair. Extensive experimental results including those on real images are presented, and they show that our method is more robust than those using position constraints alone.
For both 3-D reconstruction and prediction of image coordinates, cameras can be calibrated implicitly without involving their physical parameters. the authors present a two-plane method for such a complete calibration...
详细信息
For both 3-D reconstruction and prediction of image coordinates, cameras can be calibrated implicitly without involving their physical parameters. the authors present a two-plane method for such a complete calibration, which models all kinds of lens distortions. First, the modeling is done in a general case without imposing the pinhole constraint. Epipolar curves considering lens distortions are introduced and are found in a closed form. then, a set of constraints of perspectivity is derived to constrain the modeling process. Withthese constraints, the camera physical parameters can be related directly to the modeling parameters. Extensive experimental comparisons of the methods withthe classic photogrammetric method and Tsai's method relating to the aspects of 3-D measurement, the effect of the number of calibration points, and the prediction of image coordinates, are made using real images from 15 different depth values.< >
Recent work by Becker and Hinton (Becker and Hinton, 1992) shows a promising mechanism, based on maximizing mutual information assuming spatial coherence, by which a system can self-organize itself to learn visual abi...
Recent work by Becker and Hinton (Becker and Hinton, 1992) shows a promising mechanism, based on maximizing mutual information assuming spatial coherence, by which a system can self-organize itself to learn visual abilities such as binocular stereo. We introduce a more general criterion, based on Bayesian probability theory, and thereby demonstrate a connection to Bayesian theories of visual perception and to other organization principles for early vision (Atick and Redlich, 1990). Methods for implementation using variants of stochastic learning are described and, for the special case of linear filtering, we derive an analytic expression for the output.
In this paper, we describe the probabilistic approach to solve Poisson equations and how this method may be used to solve computervision problems. We also give a complexity analysis of this method and compare our met...
详细信息
Describes the probabilistic approach to solve Poisson equations and show this method may be used to solve computervision problems. the authors also give a complexity analysis of this method and compare it method with...
详细信息
Describes the probabilistic approach to solve Poisson equations and show this method may be used to solve computervision problems. the authors also give a complexity analysis of this method and compare it method with BEM, FEM and FDM in terms of storage and time complexity. Some examples are given and show that the authors method is better than BEM, FEM and FDM in certain aspects. Finally, the authors use the probabilistic approach for solving several computervision problems, such as shape from shading, surface interpolation and brightness based stereo vision, the experimental results show that the probabilistic approach is very effective.< >
A speech recognition method combining a neural network and fuzzy inference is proposed. this method uses the neural net as speech feature detector and fuzzy inference as a decision procedure. An unsupervised collectiv...
详细信息
A speech recognition method combining a neural network and fuzzy inference is proposed. this method uses the neural net as speech feature detector and fuzzy inference as a decision procedure. An unsupervised collective competitive learning algorithm is used to train the neural net.< >
A.P. Witkin (1981) described a method of shape-from-texture using tangent vectors of texture line segments by assuming directional isotropy of the vectors. However, the texture of natural scenes is often nonisotropic ...
详细信息
A.P. Witkin (1981) described a method of shape-from-texture using tangent vectors of texture line segments by assuming directional isotropy of the vectors. However, the texture of natural scenes is often nonisotropic with detectable bias. the authors propose a maximum likelihood estimate for recovering surface orientation from surfaces with nonisotropic texture. A priori information of the bias is used to achieve the objective. Experiments using artificial and real data show that the method is far better than Witkin's method for nonisotropic textures.< >
It is argued that traditional patternrecognition methods are inadequate for the tasks confronting computervision. In an effort to overcome their limitations, a new approach has been developed, in which the concept o...
详细信息
patternrecognition is a central topic in contemporary computer sciences, with continuously evolving topics, challenges, and methods, including machine learning, content-based image retrieval, and model- and knowledge...
详细信息
ISBN:
(数字)9783642166877
ISBN:
(纸本)9783642166860
patternrecognition is a central topic in contemporary computer sciences, with continuously evolving topics, challenges, and methods, including machine learning, content-based image retrieval, and model- and knowledge-based - proaches, just to name a few. the Iberoamerican Congress on pattern Recog- tion (CIARP) has become established as a high-quality conference, highlighting the recent evolution of the domain. these proceedings include all papers presented during the 15th edition of this conference, held in Sao Paulo, Brazil, in November 2010. As was the case for previous conferences, CIARP 2010 attracted parti- pants from around the world withthe aim of promoting and disseminating - going research on mathematical methods and computing techniques for patternrecognition, computervision, image analysis, and speech recognition, as well as their applications in such diverse areas as robotics, health, entertainment, space exploration, telecommunications, data mining, document analysis,and natural language processing and recognition, to name only a few of them. Moreover, it provided a forum for scienti?c research, experience exchange, sharing new kno- edge and increasing cooperation between research groups in patternrecognition and related areas. It is important to underline that these conferences have contributed sign- icantly to the growth of national associations for patternrecognition in the Iberoamerican region, all of them as members of the International Association for patternrecognition (IAPR).
暂无评论